# Table 13: Reserve Composition and the Safe Asset Cliff

*Connects IMF COFER reserve composition data (1999-2024) to the safe asset cliff analysis.
Tests whether demographics of reserve-currency issuers predict their currency's global reserve share.*

## Test 1: Issuer Demographics and Currency Reserve Share

| Currency | Issuer | Share 1999 | Share 2024 | OADR 1999 | OADR 2024 | Corr(level) | Corr(Δ) |
|:--|:--|--:|--:|--:|--:|--:|--:|
| USD | USA | 71.0 | 58.5 | 18.5 | 27.3 | -0.670 | 0.193 |
| EUR | EMU | 17.9 | 18.9 | 21.4 | 32.3 | -0.383 | -0.284 |
| JPY | JPN | 6.4 | 5.6 | 24.1 | 50.5 | -0.127 | -0.163 |
| GBP | GBR | 2.9 | 4.6 | 24.6 | 30.6 | 0.272 | -0.326 |
| CNY | CHN | — | 2.1 | 9.9 | 21.0 | 0.746 | 0.487 |
| AUD | AUS | — | 2.0 | 18.3 | 27.2 | 0.787 | 0.322 |
| CAD | CAN | — | 2.8 | 18.2 | 30.0 | 0.569 | 0.477 |
| CHF | CHE | 0.2 | 0.2 | 22.5 | 30.4 | -0.435 | 0.119 |

## Test 2: Panel Regression (currency × year)

| Variable | β | SE | p | N | R² |
|:--|--:|--:|--:|--:|--:|
| old_dep | -9.217 | 11.804 | 0.436 | 165 | 0.005 |
| Z₁ | -3.406 | 2.599 | 0.192 | 165 | — |

## Test 3: Reserve Concentration Over Time

| Year | USD | EUR | JPY | GBP | CNY | HHI | Mean OADR |
|:--|--:|--:|--:|--:|--:|--:|--:|
| 1999 | 71.0 | 17.9 | 6.4 | 2.9 | — | 5412 | 19.7 |
| 2005 | 64.9 | 21.9 | 4.8 | 4.7 | — | 4735 | 20.9 |
| 2010 | 61.9 | 22.9 | 4.1 | 4.8 | — | 4401 | 22.6 |
| 2015 | 64.2 | 19.4 | 3.9 | 5.2 | — | 4542 | 25.7 |
| 2020 | 59.5 | 20.6 | 6.0 | 4.8 | 2.3 | 4039 | 28.7 |
| 2024 | 58.5 | 18.9 | 5.6 | 4.6 | 2.1 | 3851 | 31.2 |

## Test 4: Safe Asset Cliff Scenarios

*Assumes freed reserve share reallocates proportionally to remaining currencies.*

| Scenario | Currencies lost | Freed share | USD new | HHI change |
|:--|:--|--:|--:|--:|
| baseline_2024 | — | 0.0% | 58.5% | +0% |
| moderate_2040 | JPY | 5.6% | 62.2% | +12% |
| severe_2050 | JPY, GBP | 10.2% | 65.6% | +24% |
| cliff_2054 | JPY, GBP, EUR | 29.1% | 84.5% | +86% |

## Test 5: Historical Downgrades and Reserve Share Changes

| Event | Currency | Share pre | Share post | Δ (pp) |
|:--|:--|--:|--:|--:|
| Japan downgraded from AAA (2001) | JPY | 5.4 | 5.6 | +0.2 |
| Japan downgraded to A+ (2015) | JPY | 4.1 | 3.9 | -0.2 |
| US downgraded from AAA (Aug 2011) | USD | 62.5 | 61.7 | -0.8 |
| UK downgraded from AAA (Jun 2016) | GBP | 4.7 | 4.7 | +0.1 |
| Eurozone crisis: GRC, IRL, PRT, ESP, ITA downgrades | EUR | 21.7 | 20.5 | -1.3 |

## Synthesis

1. **Issuer demographics and reserve shares**: Level correlations capture 
   common trends (aging + dollar decline), but first-difference correlations 
   test the incremental relationship.

2. **Historical downgrades**: Reserve shares are sticky — even large rating 
   events produce modest reallocation, consistent with reserve inertia 
   (Goldberg & Hannaoui 2026).

3. **Cliff scenarios**: The safe asset cliff (24→13.5 issuers) would increase 
   reserve concentration significantly, especially if EUR-zone safe issuers 
   (France) lose safe status.

---
*Generated: March 8, 2026. Data: IMF COFER (1999-2024), cliff_panel.csv.*